Loading...
Loading...
Contributors:
Dataset URI:
10.6084/m9.figshare.34865406.v2CataPair is a paired dataset containing 2,000 paired macular optical coherence tomography (OCT) B-scans from 98 patients imaged before and after cataract surgery at Peking University First Hospital, Beijing, China. The 4,000 original exported images were acquired with a Heidelberg SPECTRALIS OCT system during routine clinical care. Images are provided as 8-bit grayscale PNG files with a height of 496 pixels and a width of 768 pixels.Preoperative OCT was acquired within one month before surgery, and postoperative OCT within one month after surgery. A retinal specialist reviewed postoperative images to exclude surgery-related retinal structural changes and assess the stability of pre-existing retinal disease. Direct patient identifiers were removed.Each pair includes the original preoperative and postoperative images, a postoperative image registered to the preoperative coordinate system, model-derived retinal masks and boundaries, sampling-validity and evaluation masks, and saved transformation parameters. CSV manifests link these files to coded patient and scan identifiers, scan-level normal/abnormal labels, and patient-level training, validation and test subsets. The labels cover 1,774 normal and 226 abnormal pairs. The three subsets contain 67, 10 and 21 patients, respectively, contributing 1,367, 198 and 435 pairs.The dataset supports research on cataract-related OCT image restoration, registration between visits, and retinal structure analysis under changes in image quality. Registered postoperative images serve as longitudinal pseudo-references; masks and boundaries are model-derived measurements rather than manual clinical annotations. The archive includes a README, a data dictionary and a loading example. Registration and reconstruction code is available at https://github.com/citrusPericarp/paired-oct-registration.
Keyword(s):
CataPair; cataract surgery; image registration; image restoration; longitudinal imaging.
MeSH Term(s):
Grants and Funding:
Associated Publication(s):
Supplementary Data:
No Supplementary Data Available
Related Dataset(s):
No Related Dataset Data Available
The National Library of Medicine (NLM) provides information about biomedical datasets. Inclusion on an NLM Web page does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: NLM Web Policies (nih.gov)